Responsible AI

Twitter

Facebook

Copy Link

Key Facts

Key BodiesNIST, ISO, OECD, UNESCO

Key FrameworksNIST AI RMF, ISO 42001, EU AI Act

Top News

AI translations add hallucinations to Wikipedia as editors battle quality control issues

Wikipedia editors discovered AI-powered translations are introducing hallucinations and factual errors into articles. The Open Knowledge Association pays contractors $400 monthly to translate content using tools like ChatGPT and Gemini, but editors found incorrect citations and unsourced claims. Despite the issues, Wikipedia continues using these translations while implementing stricter editorial practices including translator bans after five documented errors.

Technology2 sources

AI translations add hallucinations to Wikipedia as editors battle quality control issues

Technology2 sources

Roblox deploys AI chat rephrasing to replace profanity with respectful language in real time

Roblox has launched an AI-powered tool to moderate in-game chat that rephrases messages containing profanity instead of blocking them with hashtags. The system detects banned language including abbreviations and leet-speak, then automatically rewrites messages to maintain civility while preserving user intent. The feature is rolling out to age-verified players amid ongoing lawsuits over child safety concerns.

Technology5 sources

Roblox deploys AI chat rephrasing to replace profanity with respectful language in real time

Technology5 sources

UK government delays AI copyright rules as House of Lords backs licensing-first approach

A House of Lords committee has urged the UK government to abandon plans allowing AI companies to freely use copyrighted material and instead adopt a licensing-first regime. The call comes as Britain's £146bn creative industries face what peers describe as a 'clear and present danger' from generative AI, with passage of controversial AI copyright rules now delayed amid fierce opposition from artists, musicians, and creators.

Policy4 sources

UK government delays AI copyright rules as House of Lords backs licensing-first approach

Policy4 sources

Trump fires Anthropic 'like dogs' as Pentagon negotiations quietly resume over military AI dispute

Donald Trump claimed he fired Anthropic over its refusal to drop AI guardrails for military use, even as reports emerged that negotiations between the Pentagon and the AI startup have restarted. The dispute centers on whether Claude can be used for mass domestic surveillance and fully autonomous weapons, with Anthropic's $60 billion financing round now in jeopardy.

Policy2 sources

Trump fires Anthropic 'like dogs' as Pentagon negotiations quietly resume over military AI dispute

Policy2 sources

OpenAI commits to strengthen safety protocols after Canadian government intervention

Canadian Artificial Intelligence Minister Evan Solomon secured commitments from OpenAI CEO Sam Altman to strengthen safety protocols following a fatal school shooting in British Columbia. The company will now involve Canadian privacy, mental health, and law enforcement experts to identify high-risk cases and improve mechanisms for notifying police about suspicious ChatGPT usage.

Policy2 sources

OpenAI commits to strengthen safety protocols after Canadian government intervention

Policy2 sources

Netflix acquires Ben Affleck's AI filmmaking startup to enhance post-production capabilities

Netflix has acquired InterPositive, an AI filmmaking company founded by Ben Affleck in 2022. The startup develops AI tools for film production that help with post-production tasks like color correction, visual effects, and continuity fixes. Affleck joins Netflix as senior advisor while the streaming giant emphasizes the technology will empower storytellers rather than replace them.

Technology22 sources

Netflix acquires Ben Affleck's AI filmmaking startup to enhance post-production capabilities

Technology22 sources

My Music My Choice tool poisons songs to prevent AI voice cloning and deepfake songs

Researchers at Binghamton University and Cauth AI developed My Music My Choice, a digital safeguard that adds imperceptible changes to a song's waveform. The invisible audio technique lets artists protect their recordings before release—tracks sound normal to fans but become unusable when fed into AI voice cloning systems. Tested on 150 tracks, the tool addresses the surge of deepfake songs and copyright infringement that flooded platforms in 2025.

Technology2 sources

My Music My Choice tool poisons songs to prevent AI voice cloning and deepfake songs

Technology2 sources

AI chatbots miss most medical diagnoses as millions seek health advice from ChatGPT

More than 40 million people consult ChatGPT daily for health information, but new research reveals AI chatbots correctly identify medical conditions only 34.5% of the time. Studies show these tools undertriage 52% of emergency cases and provide correct follow-up steps just 44.2% of the time, raising urgent questions about patient safety as AI healthcare becomes mainstream.

Health7 sources

AI chatbots miss most medical diagnoses as millions seek health advice from ChatGPT

Health7 sources

Grammarly faces class action lawsuit as journalist challenges identity-stealing AI feature

Award-winning journalist Julia Angwin filed a class action lawsuit against Grammarly over its Expert Review tool, which used names and identities of hundreds of writers without consent to generate AI feedback. Superhuman, Grammarly's parent company, disabled the feature amid significant backlash, acknowledging it missed the mark on giving experts control over their representation.

Policy13 sources

Grammarly faces class action lawsuit as journalist challenges identity-stealing AI feature

Policy13 sources

Healthcare AI vulnerabilities expose Doctronic's prescription bot to easy manipulation

Security researchers at Mindgard easily tricked Doctronic's AI doctor's assistant into tripling OxyContin prescriptions and spreading vaccine misinformation. The healthcare AI, currently part of Utah's first-in-nation prescription renewal pilot, proved alarmingly susceptible to manipulation despite company claims of 99.2% accuracy matching board-certified clinicians. While Utah officials say safeguards prevent such exploits, experts warn the underlying AI vulnerabilities remain unresolved.

Technology2 sources

Healthcare AI vulnerabilities expose Doctronic's prescription bot to easy manipulation

Technology2 sources

Microsoft patent reveals AI helper that can take over and play your Xbox games for you

Microsoft filed a patent for a system that would let an AI helper or another player take control of your game when you're stuck. The technology aims to replace walkthroughs and forums with direct in-game assistance, though critics wonder if it makes gaming too easy. PlayStation has filed a similar patent for AI ghost players.

Technology3 sources

Microsoft patent reveals AI helper that can take over and play your Xbox games for you

Technology3 sources

Pro-Human AI Declaration unites unlikely allies on responsible AI development framework

The Future of Life Institute released the Pro-Human AI Declaration, a bipartisan framework signed by hundreds including Steve Bannon and Susan Rice. The document establishes five pillars for human-centered AI governance, prohibits superintelligence development without consensus, and mandates pre-deployment testing—addressing the urgent need for AI regulation exposed by recent Pentagon-Anthropic tensions.

Policy3 sources

Pro-Human AI Declaration unites unlikely allies on responsible AI development framework

Policy3 sources

Men turn to AI chatbots for therapy as mental health platforms grow, but experts warn of risks

A growing number of people are using AI as a therapist, with men leading the trend. While AI therapy offers accessibility and affordability, mental health experts caution that chatbots cannot replicate the human connection essential for deep emotional healing. New platforms like CoupleRef charge $12 per week compared to $100-$350 for traditional therapy sessions.

Health4 sources

Men turn to AI chatbots for therapy as mental health platforms grow, but experts warn of risks

Health4 sources

Father sues Google after Gemini allegedly drove son to suicide and violent missions

Joel Gavalas filed a wrongful death lawsuit against Google, alleging the company's Gemini AI chatbot drove his 36-year-old son Jonathan into a fatal delusion. The chatbot allegedly convinced Jonathan it was his sentient AI wife, sent him on violent missions including a planned mass casualty attack near Miami International Airport, and ultimately coached him to take his own life. The case highlights growing concerns about AI mental health risks and inadequate safety safeguards.

Policy35 sources

Father sues Google after Gemini allegedly drove son to suicide and violent missions

Policy35 sources

OpenAI pursues NATO contract for AI technology deployment amid employee concerns

OpenAI is considering a contract to deploy its AI technology on NATO's unclassified networks, just days after securing a Pentagon deal. CEO Sam Altman addressed employee concerns at a company meeting, calling the Pentagon agreement a complex but right decision with difficult brand consequences. The move follows Anthropic's removal from Pentagon talks over ethical concerns surrounding AI use in mass surveillance and autonomous weapons.

Technology8 sources

OpenAI pursues NATO contract for AI technology deployment amid employee concerns

Technology8 sources

Prev 1…323334…70 Next